Can computer science help us live longer? It is a complex question that requires an understanding of our anticipated lifespan and how the health of our body at this moment compares to the average person our age, what some would call calculating our biological age. What computer science stands to help with is understanding how better to pinpoint that biological age and what factors may play into holding off the aging process. Guest Andy Lee, of NeuroInitiative and Vincere Biosciences, speaks with host Angelo Kastroulis about this hard problem of predicting our biological age and potentially reversing it.
Can computer science help us live longer? It is a complex question that requires an understanding of our anticipated lifespan and how the health of our body at this moment compares to the average person our age, what some would call calculating our biological age. What computer science stands to help with is understanding how better to pinpoint that biological age and what factors may play into holding off the aging process. Guest Andy Lee, of NeuroInitiative and Vincere Biosciences, speaks with host Angelo Kastroulis about this hard problem of predicting our biological age and potentially reversing it.
Angelo begins with a discussion on some of the classical age models that have previously looked at the question of determining one’s biological age. He talks about classic papers by Horvath and Fahy that have changed the way scientists think about aging. He introduced the idea of epigenetics, or the study of the changes in living things that are caused by the medication of the way genes are expressed rather that the more classical mode of altering the genetic code itself.
With his Guest Andy Lee, founding CTO of NeuroInitiative and COO of Vincere Biosciences, Angelo dives deeper into DNA methylation signatures and the patterns that we can look at to begin to determine someone’s biological age. Andy describes how computer science and neural networks are modernizing these determinations and what that means for improving our longevity.
The pair note the challenges posed by the sheer volume of genetic data and what advances in data science can make in our ability to push this area forward, including therapeutics for diseases such as Parkinson’s. They talk about how computer science is allowing us to have transformative information brought to us so that then we can intervene and act on it.
About this Episode’s Guest
Andy Lee is Co-Founder, Director, and CTO of NeuroInitiative, where he is co-inventor on multiple granted and pending patents surrounding the SEED simulation platform, as well as COO at Vincere Biosciences, Inc., a Cambridge, MA, company developing disease-modifying therapies for Parkinson's disease. Previously, Andy was VP of Engineering at Black Knight through F500 acquisition, spin-out, and IPO. He has led teams of over 100 members and continues to actively code to create new data-driven solutions. You can find out more on Twitter (@Andy_D_Lee) and LinkedIn.
Citations
Fahy, GM, Brooke, RT, Watson, JP, et al. Reversal of epigenetic aging and immunosenescent trends in humans. Aging Cell. 2019; 18:e13028. https://doi.org/10.1111/acel.13028
Horvath S. (2013). DNA methylation age of human tissues and cell types. Genome biology, 14(10), R115.https://doi.org/10.1186/gb-2013-14-10-r115
Johnson, A.A., Shokhirev, M., and Shoshitaishvili, B. (August 2019). Revamping the Evolutionary Theories of Aging. Ageing Research Reviews. 55. Doi: 10.1016/j.arr.2019.100947.
The Matt Walker Podcast. https://sleepdiplomat.com/podcast
Further Reading
Handbook of Epigenetics, 2nd Edition
Andy Lee on Twitter and LinkedIn
About the Host
Angelo Kastroulis is an award-winning technologist, inventor, entrepreneur, speaker, data scientist, and author best known for his high-performance computing and Health IT experience. He is the principal consultant, lead architect, and owner of Carrera Group, a consulting firm specializing in software modernization, event streaming (Kafka), big data, analytics (Spark, elastic Search, and Graph), and high-performance software development on many technical stacks (Java, .net, Scala, C++, and Rust). A Data Scientist at heart, trained at the Harvard Data Systems Lab, Angelo enjoys a research-driven approach to creating powerful, massively scalable applications and innovating new methods for superior performance. He loves to educate, discover, then see the knowledge through to practical implementation.
Host:Angelo Kastroulis
Executive Producer: Kerri Patterson; Producer: Leslie Jennings Rowley; Audio Engineer: Ryan Thompson; Communications Strategist: Albert Perrotta
Music: All Things Grow by Oliver Worth
© 2021, Carrrera Group
Angelo Kastroulis: I am 47 years old, but what if I could turn back the clock to be 44 and a half years old? Steve Horvath and Greg Fahy showed you could do just that in their research papers, on reversing the epigenetic aging. In today's podcast. We're going to talk about that, the difference between the chronological age and your biological age, and how to reverse it.
I'm your host Angelo Kastroulis, and this is Counting Sand.
Can computer science help us live longer? Well, first we have to kind of figure out what is a human lifespan and what causes our lifespans to differ. In 2013, Steve Horvath wrote a paper entitled “DNA methylation age of human tissues and cell types.” This is a scientific paper where he presented a new and emerging model in epigenetics. What does that have to do with computer science? Computer science plays a really important part.
Our age can be thought of chronologically. In other words, how many years have we been alive and have gone around the sun, but that doesn't necessarily normalize what is a 50 year old person. For example, I'm 47. Do I have the average life expectancy of someone my age? Am I actually a 47 year old and a 47 year old's body? Or am I in 47 year old in a 50 year old’s body. That's where our biological age comes into it. So, how do you figure that out? Well, we could try to take some factors like health and the types of maybe biomarkers that we might have in our blood and things like that. There are classical age models that we've thought of for many years. For example, there's a disposable Soma model that says that effectively, we have this tension in our body between repairing the body and reproducing. And so that we have a finite amount of resources and there's a trade-off between the two. And at some point, as we grow older, that tipping point changes. And that's why we grow older. The Horvath 2013 paper was a very important paper because it changed the way that we think about aging. It added some new kinds of models to augment the classical kind of thinking. These new kinds of models are called epigenetics.
So the big idea here is that changes to an organism can be defined not just by altering the DNA, the genetic code itself, but rather the expression of the genes. What makes that really interesting as you can imagine how hard it is to change DNA, but you can change the way it is expressed. And that means that we are able to affect change to our own lives.
If computer science can help us to understand how we can collect and understand this data, what is the effect and interventions we have to be able to change this expression? Computer science can help us to live better and longer lives. So in some of the classical aging theories and this dichotomy, there is with a finite number of resources and there's always tension between them.
There's not a lot you can do about that. However, the proposition raised by epigenetics is that these processes can be reversible and they’re reversible, because these phenomena are controlled, not just through gene mutations, but by other processes that happen, they can range from things like DNA methylation, which we'll talk about a little bit, or prions. Prions is just a generic term used to describe some mysterious infectious agent that is traversing through our body. In the Horvath 2013 paper, he goes on to show a very, very strong correlation between our chronological age and the amount of methylation in our DNA. And we'll talk about what that is in a minute.
By creating such a strong tie, and we now have a way to manage it. In a subsequent paper that he published in 2019, he actually showed that an intervention was able to reduce the biological age so that the person effectively went back in time.
I am joined by my close friend and colleague, Andy Lee, CTO of NeuroInitiative and COO of in Vincere Biosciences. Andy focuses on using technology to speed up pharmaceutical development. They've built a patented biology simulation platform and a suite of supporting tools that led to a pipeline of some small molecules being developed to stop Parkinson's disease and other age-related disorders. But they're also working on lots of other interesting stuff like biological age. Andy and I have known each other personally and professionally for a very long time. And in the interest of full disclosure, I'm an investor in Vincere Biosciences. I am really excited to be able to talk to him about this hard problem of predicting our biological age. Andy, welcome to the podcast.
Andy Lee: Thanks for having me, Angelo. Yeah, this goes back quite a while. I think it's been well over 10 years since before either of us were doing healthcare stuff at all.
Angelo Kastroulis: So we've known each other forever and it's been kind of fun to watch the evolution of both of our initiatives and companies growing up from ideas. You were incubating back in IBM's incubation lab. Right?
Andy Lee: We, we were, we started in their startup program and they actually funded our GPU clusters on a soft layer. Back then they were one of the only platforms that actually had an Nvidia GPU's in a cloud hosted platform. And then, since then it's now available on Azure, Google and AWS.
Angelo Kastroulis: What's funny is that the time I was just getting into GPS as well, for a completely different reason, I was studying them at Harvard state and system lab to see how we can push database computations onto GPU so that databases could under the covers take advantage of that kind of thing in some computations. So even though we haven't seen each other in a while, we've been very orthogonal. You're in Cambridge now, right?.
Andy Lee: We are. Yeah. You know, we, we started there in Jacksonville. There's, there's quite a strong software engineering community there that was helpful to tap into and some good people to bounce ideas off of to kind of get that stuff going. And it's a little more, you know, internet, community friendly, I suppose getting into the real world therapeutics and partnering with pharmaceutical companies and, and talking to venture capital, Cambridge is just the center of the universe. So that was really a no brainer to move up here when we spun off the therapeutics piece.
Angelo Kastroulis: But I don't want to paint the picture we haven't seen each other in forever. So when I was up at Harvard and we were able to hang out in Cambridge sometimes.
Andy Lee: That was just, yeah. And a small world kind of funny that I, several times just bumping into your own town here, which is kinda neat to have that it is very much a, I guess, a small world and also how interconnected these, these, communities of biology and technology are becoming.
Angelo Kastroulis: Okay. So today's topic is the hard problem of computing your biological age. What exactly is that? Well, let's start really by asking what's our chronological age. That one's a bit easier to answer. It's basically just the number of years we've been alive. The number of times we've gone around the sun. But that doesn't really tell us what that means. We can compare ourselves to other people who are the same age and then kind of get an idea that we might all have a similar life span, but we won't. And we could try then to take a statistical average of how long do people in maybe certain areas live. Just try to understand how to compare ourselves, but it's a really weak comparison.
What might be more interesting is not to say how many years have I been alive, but maybe to say how healthy is my body? In comparison to maybe the length of my lifespan, where am I in this lifespan? That's a really, really hard question to answer. But if we step back for a second and think of it a little bit differently, what if we were to say that your body is a certain age and that's not necessarily tied to the number of years you've been alive?
So say that I'm 47 years old, but I have the body of a 45 year old. That's positive. That means I'm in better shape than maybe what I would have been if I would have compared myself to just numerically to other people. Or if I have the body of a 50 year old, but I'm actually only 47. That means I need to stop and take some action to be able to correct that so that my body's health goes back in line where it should be. That's a pretty important thing to know, especially if we can have some kind of intervention that will bring us back in line. But the question you might be asking yourself is could I then even improve it further so that effectively moving back the clock. So instead of having the body of a 47 year old and being 47, I can actually maybe do better and improve it and push back time a little further. We're going to get into that research here in just a little bit.
Andy, tell us a little bit about this hard problem of computing and biological age.
Andy Lee: It's one of the things that's really exciting that has just come to really be understood in the last five or 10 years is that biological age and chronological age are not the same thing they can diverge and you can measure biological age independent. And excitingly, you can actually modify biological age. So this idea that we all degenerate toward death at some point on a kind of a fixed clock, everybody, you know, there's median life expectancy, but then you you're plus, or minus 20 years, and then that's about it, you know, we've now been able to see in the laboratory, the ability to as much as triple the lifespan of, of animals sometimes, and in some cases doing that with a single genetic mutation that you can change something and extend the lifespan. So now, knowing that those processes are extendable, it is…the next question is like, how do we know that we're doing that in humans? If you know, you can't change it, if you can't measure it. Right? So we really, I try to comment life sciences with an engineering perspective and it is like, okay, if we're going to do this, let's quantify it. Let's figure out the specific steps that can do it. How do we track it and improve it? So that's where measuring biological age becomes impatient.
Angelo Kastroulis : Andy. That's a big concept. If we could actually intervene and reverse it, what scientific research is out there that kind of supports this way of thinking?
Andy Lee: The first major breakthrough, and this was Steve Horvath wrote a paper showing that he could reliably predict a biological age using DNA methylation signatures. DNA encodes the proteins that make up the whole system of life. And throughout time methyl tags added to the DNA at certain places that change in a predictable pattern. So they actually, in that first system just did simple linear regression models against that to identify 370 or so sites that would change together in a pattern to be able to predict the age. So now seeing that platform is, I think it's plus or minus three years, it's really opened up a desire to make those predictions more accurate right now. So we're trying to get more specific and see if there are better signatures using some deep neural network models to interrogate those methylations sets.
Angelo Kastroulis: So that's where computer science fits in all this computer science and biology come together to see if we can accurately measure our biological age by looking into our DNA. You mentioned DNA methods. I think it might be worthwhile for our listeners to take a small digression and talk about what methylation is. Methylation is where we take a methyl groups and we bind them to our DNA. Now we want to be careful. We don't want to change the DNA because that won't be anything that we can measure and understand. But if we can bind methyl groups to DNA, that has a specific effect on our DNA. It affects the DNA is transcribability. So as it turns out over time, the effect of these methyl groups and the rate of transcription changes with our age and it's predictable and that forms the basis of this epigenetic clock that we can use to measure, to figure out what is our biological. Where you are in this rate of change is how old you are and then we can accurately try to understand your age. At least that's the hope. Andy, we talked about a 2019 paper that Horvath and Fahy wrote where there was a big breakthrough in this. Can we talk about that for a second?
Andy Lee: Yeah, Greg Fahy is the first author and actually had the opportunity to see him speak in a longevity webcast and got to ask him a couple of questions, which was pretty exciting. He started a study using growth hormone as a treatment to regenerate the thymus. And from that reverse the biological age as measured through this epigenetic clock. And he actually coupled that with multiple others, there are now three or four different methylation clocks that people have come up with different sets of markers that predict, and they're all relatively comparable as far as accuracy. And then another approach looking at plasma and some markers in the blood that are other than the DNA methylation that can do the same thing. So through this, they were able to show reversal of age. So people would come in, you calculate their baseline by looking at biological age minus their chronological age to figure out are they above or below to coming in and then track them over a year of treatment and see, are they further above or below their chronological age at the end. And they were on average able to shift people, reduce their biological age by about two and a half years over a one-year treatment plan. So really turning back the clock and making people younger throughout this pretty exciting concept that you may be able to take a treatment and not the younger.
Angelo Kastroulis : Andy, you mentioned there were different kinds of epigenetic clock. So there's not really a gold standard to this, not one magical clock that we can all just say, that's it. There's some of the others measure the length of 10 telomeres and try to make some determination back on that. I know some use biomarkers. Those show a bit of promise. Can you talk about that for just a second?
Andy Lee: Yeah. And you know, some of these are things that we don't even think about in terms of longevity. Like you go and get your annual checkup with cholesterol and you see your HDL and LDL numbers that start to go up as you age, and then you take statins or other diet changes to try to bring that back down. Well, that pattern of going up as you age is an effect of your biological age changing. So that itself is a marker that you can start to change. PSA levels that get measured for prostate cancer. You know, those are something that start to climb as, as people age. So all of these things that you can start to track, you can turn back, some are even quite a bit easier.
Angelo Kastroulis: Okay. So aside from taking DNA itself, there are other models that rely on things outside of methylation like our skin elasticity. Andy you mentioned PSA levels, cholesterol. So while we can do all of that and create different kinds of models, how do we put all these things together to be able to compute these things? Again, taking it back to computer science, one technique, and you mentioned this, this machine learning, artificial neural networks, which are really good at finding patterns in very, very big sets of data. Andy, how are you using artificial neural networks to be able to, to do that?
Andy Lee: One of the interesting things we're playing with our neural networks for image recognition. So off of an image analysis of a face you can get within, some people are claiming models with plus or minus two years up to like eight years seems to be pretty doable, but most models, so comparable to some of the biological patterns, just a face recognition trained on a large set of age-matched faces can then detect a person's age off of the face. And some of these things we all intuitively know, you get more wrinkles, you get less hair. The neural nets are really good at picking up those types of signals through a large enough dataset.
Angelo Kastroulis: But there are challenges to creating artificial neural networks. For example, they have to be trained by giving them images. How does it know that that picture is a 47 year old or a 20 year old someone would have to label that picture, right?
Andy Lee: Yeah. And with most deep learning projects, I think the first challenge is having enough accurately annotated data to train that. It’s challenging, especially when people look different in the morning versus the night, whether they're got makeup on or not in different types of lighting situations, whether they're looking straight on. You know, we've noticed things like if you look up, it will predict you as younger. If you look down that will predict you as older.
Angelo Kastroulis: One of the ways to counteract that is maybe to have more variability in the image, but that's kind of a double-edged sword.
Andy Lee: Yeah. As you know, there are more points that can be measured than are being measured in and there's variability. From individual to individual and from tissue type to tissue type and from cell to cell. And so the more we pool all of that data, the less useful it becomes because you start to just get such wide error bars that you can't really get as precise as we'd like to for diagnostics or tracking decisions. But within any one of those, you start to see companies like 10X Genomics over the last few years have been innovating in single cell technologies. And so now they have sequencers that you can take samples through and it flows it through a device that will sequence the DNA of each cell. So you can get single cell results in both looking to see if there are mutations to reach of those looking at you can get single cell transcriptomic levels to see if there are different levels of MRNA. So beyond the template that the DNA provides, based on certain other factors, you'll get different quantities of RNA and of protein that they created. So those things all drive the system of the way the cell behaves differently. And then with the methylation we've seen in the last few years that we've been looking at this since Horvath's first paper, the chips that can quantify the methylation changes looked at about 400,000 sites. And now there are billions of sites that are available, the newest chips that are growing. So we have just this much larger data set, much wider data set to work with. And in order to then overcome the variability from individual-individual, we're looking at hundreds of thousands, millions if we can get them of separate human samples that we want to then analyze, pulling that data together, normalizing it, getting it all annotated similarly, and then getting it into a matrix that you can run a deep learning model against becomes the difficult thing. Just, just an ETL pipeline challenge in itself becomes difficult.
Angelo Kastroulis: Andy, you touched on a really important point of how big the data is. You mentioned an ETL pipeline challenge. It's true. That that is the case. Neural networks love a lot of data. So a lot of humans with a lot of genetic code that we can produce to train on is great. But even if we encode the genome, say we can take a gene and we could just squish it into just a few bites of data, if we could pack it in as tightly as we could, you're looking at gigabytes for one person's genome. But as you pointed out that, that in some cells, a DNA sequence in one cell will be different than the DNA sequence in another. Let's just say we have a hundred trillion cells in our body. I mean, you're talking about zettabytes of data and if you're talking about millions or billions of people, you can't even move that kind of data. Let alone, I wonder if there's even enough storage on a planet to store.
Andy Lee: What's exciting though, is we're starting to get some of that data available if you'd look at DNA sequencing and just looking for mutations within that really became a plausible thing to look at in what was it, the late nineties that they did the, the human genome project and did that first 10 years or so, and billions of dollars to get a first full sequencing. And now the Broad Institute down the street is sequencing about one genome, every 10 minutes there. What you take millions is now just a few dollars and 10 minutes of time. And they're just like flowing it through and kicking out all these data sets and then sharing them in a way that can be leveraged by outside researchers, which is pretty amazing. We're actually in a similar inflection point, I think with methylation right now. And I think that's going to open a lot of new doors that these modifications drive behavior changes within the system. Right now, we're still at that point where it costs about 500 to a thousand dollars to sequence the methylation sites for a single individual. And that takes weeks to the process given the types of providers that can do it.
Angelo Kastroulis : And I think it's interesting that you pointed out that we don't need to sequence the entire DNA genome in order to have the data that we need. We can just try to get to the methylation sites and just sequence that then that simplifies the problem.
In future podcasts, we'll talk about more opportunities for us to kind of simplify the problem. In fact, that's a great computer science general topic is problem simplification. What can we boil it down to something we can solve? And then by solving the problem in lots of different pieces, you can end up solving the whole problem, just like Archimedes did.
Andy, let's talk about the idea that you mentioned about modifying our behavior, at least influencing it, because I think if we're all armed with the right kind of information and we knew what kinds of interventions we could take, and some of them are surprisingly simple, that would really have a big impact on our lives. Can we talk about that for a second?
Andy Lee: Whereas we've been starting to kind of talk with people who may be interested in having their biological age measured. I think one of the key things people are looking for is tips and guidance on how they can improve like telling me, I've got 40 years left is one thing telling me how I can change that timeline is entirely different than thus far, I think all of the known interventions are things that are pretty healthy and good ideas to do anyway. Caloric restriction, Mediterranean diet, and getting exercise and getting good sleep. Right? These are things that if we all want to operate at peak performance, we should be doing anyway. And so I think the biological age is just another helper to kind of guide you toward like, yeah, you need to do that. And I, I think coming back to measuring things, a lot of people started getting out and working out more, or at least walking more when they got Fitbits going around and people could actually see, wow, I've sat around all day. And so if we're talking about knowing that we have the ability to give people actionable guidance that will make them healthier and live longer and feel better. I think there's an ethical requirement to do those things, right? To sit on that information and not get it out to people. I think that becomes problematic.
Angelo Kastroulis: So then biological age measurement becomes a way for us to understand and quantify these things, right? It's not just trying a new diet and seeing what happens. You can quantify it by weight loss. But looking at your age is a little harder. So now if we can quantify it, we can see how these interventions can help us, but they can also serve as a wake up call. And if we have our biological age and it's so far out of sync in the wrong direction, that's a good wake up call for us.
Andy Lee: Yeah. Absolutely. And now that we're getting more involved in therapeutic development and our primary focus is around Parkinson's disease, but we now know that there are ways to figure out the biology as a system, there are proteins that interact in certain ways to drive behavior, and that changes over time and in response to environmental factors and genetics, and you can manipulate that. You can manipulate that through gene therapy to introduce or remove genetic factors. The MRNA vaccines that we're all happy to have right now that introduced the template for antibodies to protect us from disease or small molecules that you can take that bind to receptors and change the behavior. So now that we know that the biological age can change. We're starting to really probe for pieces that can be intervened on. Everybody wants the magic pill. And I think the magic pill is out there. And while exercise, sleep and diet feel good, they're also not something that everybody wants to do all the time. It's something you have to think about every day and. I don't mind taking vitamins that can keep my energy levels a little bit better. If I can take another one that I can measure and know I’m improving my methylation age or that I can measure my telemeres lengthening, or I can see my mitochondrial health getting more efficient. You know, I, I think that's going to be really compelling and it feels doable as a drug developer and as a part of a biology team, looking at these things in depth. I think we're, we're at the point where some of these things that we're seeing in the laboratory in animal models now are turning into humans. The study that Fahy and Horvath put, showed reversal of biological age with an intervention. They're actually now running a larger trial on that to see if it holds up in a larger cohort. So it's exciting, exciting stuff to see. And I think it's going to change the way we think about health. You know, we're already seeing it with the trackers that we have on some things. And within the next few years, we're going to have things that continuously monitor our glucose and our heart rate and our health and all of these other metabolites we don't even know to measure yet that are markers of biological age. If you can see this progressing within a one year or six month or a month long scale, it's going to be so much more exciting to track that. I'm energized to stay on an exercise diet plan because I can get on the scale and see the numbers going down a point every week or two. And I just can't do that with my aging right now. The age-related things that we look at take too long. Getting those diagnostics that allow people to have that visibility into their. I think are really going to be game changers and knowing that these new interventions work and in encouraging people to use the interventions as they become available.
Angelo Kastroulis : And there it is right there. Computer science is the glue that brings biology and biological sciences to. That is how we're able to have transformative information brought to us so that then we can intervene and act on it. Computer science not only takes that biology and helps us understand it through things like artificial neural networks or machine learning or other techniques, but it also is the glue that brings it all together and brings it directly to us, to our devices. The measurement systems and then making sense of all that data. And that is why I love computer science. And that is why I do what I do.
Andy Lee: Yeah, yeah, for sure. What I pivoted into life sciences six or seven years ago now that was one of the things that really drove it is, you know, seeing all of these techniques and technology that we were using for big data and analytics and predictions in FinTech. And then seeing the massive amounts of new data that were coming available in the life science space. And how few people in life sciences had the computer engineering backgrounds to be able to leverage those data sets. And it really has been this, this great confluence of new techniques to produce biological data, new computational approaches, like GPU's and others that have allowed us to move massive amounts of data to process them in ways that we couldn't do 10 years ago and having that come together. We're just seeing the tip of the iceberg, right. It is such early days in life science technology. I would encourage anybody who has interest in these spaces to pick a disease, pick an area, pick some specialty that you are passionate about and dive into it because the science world needs technologists and software engineers to get in, learn enough of the science to help out, to talk, to speak the language to build those solutions and getting them forward.
Angelo Kastroulis: I couldn't agree more. I think I've said this before, but we're really lucky to be alive in the period that we are. There's so much advancement, so much opportunity we can be a part of. We've talked about a lot of things and we've only touched on a few really interesting topics that we're going to get more of in the future. And we only scratch the surface of on topics like mitochondrial health and sleep. Sleep is one of those things that I've only recently become to realize is really, really important. If you hadn't had a chance to listen to Matthew Walker's podcast, I really recommend you do that.
If you want to learn more about computing your biological age, follow the work that Vincere Biosciences is doing. Andy, thanks for joining us. Thanks for taking the time to talk about how hard these problems really are, but really how much they can change our lives. If you're a computer scientist and you truly want to work on some interesting problems and you want to jump into healthcare, biology or genomics, there's all kinds of room for you for innovation here.
And thank you listener for joining us. If you want to learn more about the cool projects NeuroInitiatives and Vincere Biosciences are doing, you'll find Andy's Twitter and LinkedIn accounts along with my own in the show notes. I'm Angelo Kastroulis and this has been Counting Sand. Please take a minute to follow rate and review the show on your favorite podcast platform so that others can find us. Thank you so much for listening.